Entry Point

Train

Module to train a network using init files and a CLI

deepreg.train.build_callbacks(log_dir: str, histogram_freq: int, save_period: int) → list

Function to prepare callbacks for training.

Parameters

  • log_dir – directory of logs

  • histogram_freq – save the histogram every X epochs

  • save_period – save the checkpoint every X epochs

Returns

a list of callbacks

deepreg.train.build_config(config_path: (<class 'str'>, <class 'list'>), log_root: str, log_dir: str, ckpt_path: str) → [<class ‘dict’>, <class ‘str’>]

Function to initialise log directories, assert that checkpointed model is the right type and to parse the configuration for training

Parameters

  • config_path – list of str, path to config file

  • log_root – str, root of logs

  • log_dir – str, path to where training logs to be stored.

  • ckpt_path – str, path where model is stored.

Returns

  • config: a dictionary saving configuration

  • log_dir: the path of directory to save logs

deepreg.train.main(args=None)

Entry point for train script

deepreg.train.train(gpu: str, config_path: (<class 'str'>, <class 'list'>), gpu_allow_growth: bool, ckpt_path: str, log_dir: str, log_root: str = 'logs')

Function to train a model

Parameters

  • gpu – str, which local gpu to use to train

  • config_path – str, path to configuration set up

  • gpu_allow_growth – bool, whether or not to allocate whole GPU memory to training

  • ckpt_path – str, where to store training checkpoints

  • log_root – str, root of logs

  • log_dir – str, where to store logs in training

Predict

Module to perform predictions on data using command line interface

deepreg.predict.build_config(config_path: (<class 'str'>, <class 'list'>), log_root: str, log_dir: str, ckpt_path: str) → [<class ‘dict’>, <class ‘str’>]

Function to create new directory to log directory to store results.

Parameters

  • config_path – string or list of strings, path of configuration files

  • log_root – str, root of logs

  • log_dir – string, path to store logs.

  • ckpt_path – str, path where model is stored.

Returns

  • config, configuration dictionary

  • log_dir, path of the directory for saving outputs

deepreg.predict.build_pair_output_path(indices: list, save_dir: str) -> (<class 'str'>, <class 'str'>)

Create directory for saving the paired data

Parameters

  • indices – indices of the pair, the last one is for label

  • save_dir – directory of output

Returns

  • save_dir, str, directory for saving the moving/fixed image

  • label_dir, str, directory for saving the rest outputs

deepreg.predict.main(args=None)

Function to run in command line with argparse to predict results on data for a given model

deepreg.predict.predict(gpu: str, gpu_allow_growth: bool, ckpt_path: str, mode: str, batch_size: int, log_dir: str, sample_label: str, config_path: (<class 'str'>, <class 'list'>), save_nifti: bool = True, save_png: bool = True, log_root: str = 'logs')

Function to predict some metrics from the saved model and logging results.

Parameters

  • gpu – str, which env gpu to use.

  • gpu_allow_growth – bool, whether to allow gpu growth or not

  • ckpt_path – str, where model is stored, should be like log_folder/save/xxx.ckpt

  • mode – train / valid / test, to define which split of dataset to be evaluated

  • batch_size – int, batch size to perform predictions in

  • log_dir – str, path to store logs

  • sample_label – sample/all, not used

  • save_nifti – if true, outputs will be saved in nifti format

  • save_png – if true, outputs will be saved in png format

  • config_path – to overwrite the default config

deepreg.predict.predict_on_dataset(dataset: tensorflow.data.Dataset, fixed_grid_ref: tensorflow.Tensor, model: tensorflow.keras.Model, model_method: str, save_dir: str, save_nifti: bool, save_png: bool)

Function to predict results from a dataset from some model

Parameters

  • dataset – where data is stored

  • fixed_grid_ref – shape=(1, f_dim1, f_dim2, f_dim3, 3)

  • model – model to be used for prediction

  • model_method – str, ddf / dvf / affine / conditional

  • save_dir – str, path to store dir

  • save_nifti – if true, outputs will be saved in nifti format

  • save_png – if true, outputs will be saved in png format

Warp

Module to warp a image with given ddf. A CLI tool is provided.

deepreg.warp.main(args=None)

Entry point for warp script

deepreg.warp.warp(image_path: str, ddf_path: str, out_path: str)
Parameters

  • image_path – file path of the image file

  • ddf_path – file path of the ddf file

  • out_path – file path of the output